The candidate, Dr. Rafael Bejar, presents a 5-year career development plan that seeks to characterize the broad range of mutations present in patients with myelodysplastic syndromes while establishing an academic career as a physician scientist in the field of hematology. The specific aims of this proposal are to: (1) study a clinically annotated cohort of 439 bone marrow samples from patients with MDS to determine if the mutations they carry are associated with common clinical features of MDS and overall patient survival, (2) perform a pooled shRNA screen targeting each of the genes from the commonly deleted regions of chromosome 5q to determine which genes, when knocked down in human CD34+ hematopoietic cells, lead to clonal expansion in a stromal co-culture system, and (3) functionally characterize the mutations identified in MDS patient samples and candidate target genes from the shRNA screen to determine their effects on hematopoietic differentiation, clonal expansion, cell cycling, and apoptosis in an in vitro model system. Myelodysplastic syndromes are clonal disorders of hematopoiesis that cause inefficient blood cell production, low blood counts, and risk of progression to acute leukemia. The prognosis for patients with MDS is highly variable, with some only mildly affected while others succumb within months of diagnosis. Chromosomal abnormalities present in less than half of cases help stratify patients into risk groups, but prognosis is largely determined by clinical parameters. Patients with MDS have been shown to have acquired point mutations in the DNA of their diseased cells, however, these genetic lesions have not yet been incorporated into prognostic scoring systems. Molecular markers are needed to better classify subtypes of MDS, stratify prognostic risk in patients, and identify pathogenic mechanisms associated with the development and progression of these disorders. There is abundant evidence that genes that are somatically mutated and functional in one tumor type can play a role in other tumor types. A comprehensive analysis of known cancer genes in MDS has not been performed. In order to identify oncogenic mutations in MDS, a set of DNA primer extension/mass spectroscopic assays (collectively called OncoMap) will be used to determine which of 439 patient samples contain any of 1060 mutations in 104 known cancer genes. Tumor suppressor genes known to be mutated in MDS and other myeloid disorders will be sequenced in this cohort using the Roche/454 quantitative next-generation sequencing platform. These highly sensitive techniques will allow the detection of mutations even if samples contain a large fraction of normal cells or mutations are present in only a small subclone of diseased cells. With the help of collaborator Donna Neuberg and her team, a statistical analysis will be performed to determine if mutations are correlated with MDS phenotypes and the overall survival of patients even after known risk factors are considered. In collaboration with Dr. David Root from the RNAi Consortium at the Broad Institute, a pooled short hairpin RNA interference screen in human CD34+ hematopoietic cells will be performed. Genes from regions of chromosome 5q that are commonly deleted in patients with MDS will be targeted to determine which confer a clonal advantage when knocked down. After validation, gene targets identified in this screen and genes that are mutated in patient samples will be tested using in vitro human hematopoiesis assays to determine their effects on aspects of the MDS phenotype including blood cell development, apoptosis, cell- cycle changes, and clonal expansion. Dr. Bejar is well qualified to carry out the research outlined in this proposal. He has successfully completed a project of comparable complexity as part of his PhD thesis. His mentor, Dr. Benjamin Ebert, has experience studying the genetic basis of myelodysplastic syndromes and has published the result of short hairpin RNA interference screen that discovered RPS14 as a critical gene lost in patients with the 5q-minus syndrome subtype of MDS. Along with Dr. Ebert, Dr. Bejar has published an invited review on the genetic basis of MDS in Hematology/Oncology Clinics of North America and has a review on pathogenic mechanisms in MDS in progress for eventual publication in the Journal of Clinical Oncology. Dr. Bejar has recruited a team of outstanding mentors that that in addition to Dr. Ebert, include Dr. David Williams as a co-mentor, Dr. David Steensma as a clinical mentor, and Dr. Nancy Berliner, head of the Hematology division at Brigham and Women's Hospital, as a fourth member of his advisory committee. The advisory committee will meet, at minimum, every six months to ensure a successful scientific research program. Included in the career development plan are courses on biostatistics and genomic methods at the Harvard Medical School and School of Public Health. Successful completion the specific aims and career development plan outlined in this proposal will allow the candidate to advance his academic career as an independent investigator in the field of hematology.